696
INDUSTRIAL AND ENGINEERING CHEMISTRY
phenol the color is much the same as with phenol, but the test is about one half as sensitive. The color from pchlorophenol is more toward the blue, and the sensitivity is only about one fourth that of phenol. LITERATURE CITED
Am. Puhlic Health Assoc. and Am. Water Works Aesoc., New York, “8tandard Methods of Water Analysis”, 8th ed., 1936. (2) Baylis, J. R.,J. Am. Water Works Assoc., 19, 604 (1929). (1)
Vol. 16, No. 11
(3) Baylis, J. R.,Water Works and Sewage, 79,343 (1932). 14) Buswell, A. D., and Dunlop, E. c.,“Estimation Of Phenols h Water by Mems of Ultraviolet Absorption Spectra”, presented at 103rd Meeting of AMERICAN CHEMICAL SOCIETY, Memphie, Tenn. (5) Cohen, B., Gibbs, H. D., and Clark, W. M., Pub. Hculth Hepta., 39,381, 804 (1924). (6) Rensoni, L. S., J. Am. Water Work8 Assoc., 32, 1035 (1940) (7) Theriault, E.J., IND. ENO.CHEM.,21, 343 (1929). (8) Turker, I. W., J. Assoc. Oficial A g r . Chem., 25, 779 (1942! (9) Williams, R. D., IND. ENQ.CHEM.,19, 530 (1927).
Determination of the Nutritive Value of the Proteins
of Food Products H. H. MITCHELL, Animal Nutrition Division, University of Illinois, Urbana, 111. The nutritive value of food protein cannot b e accurately assessed lrom 4 determination of the amino acid content, nor from animal feeding experiments that fail to credit the protein with all its functions in the body. A study of the nitrogen economy of the animal when fed the protein to b e tested under certain essential conditions Is capable of giving 4 complete and reasonably satisfactory picture of protein utilization in the body, in digestion as well as in metabolism, The net protein value combines all this information in one @sure. Illustrations of application of the method to problems ot the storage 4nd processing of foods are given.
TvaryHE
protein content of food products has a n uncertain significance for their evaluation in nutrition, because proteins widely in the extent of their digestion in the alimentary canal, and even more so in the extent to which the end products of their digestion, largely amino acids, are available for those functions i n the body peculiar t o dietary protein. These functions are bewildering in their complexity and in their involvement in life processes, but quantitatively the construction of new protein material in growth and the maintenance of the nitrogenous integrity of the tissues already formed are by far the most important. The differences that exist in the value to the animal body of the proteins in food products, the extent to which food proteins supplement each other’s amino acid deficiencies when combined into diets, particularly the best method of supplementing the proteins of white flour, and the effect of storag? and commercial p r o m i n g on the protein value of food products in nutrition are all problems of importance to food technologists. How can these problems be most effectively tackled in the chemical or the biochemical laboratory? At the present time, three methods are being used for these purposes: ( a ) determination of the amino acid contents of foods or food proteins, (b) measurement of the ratio of gain in weight to protein intake of growing rats subsisting on rations in which the protein content is the only factor limiting growth, and (c) measurement of the gain in nitrogen to the bodies of growing rats resulting from the consumption of diets complete in all respects but protein. The latter method can be extended to mature, pregnant, or lactating animals, and to the human subject. This paper considers briefly the advan tages and disadvantages of each method for the purposes for which they are being employed. AMINO ACJD ANALYUS OF PROTEINS AND FOODS
The basis upon which this method rests is that the nutritive quality of food proteins is determined entirely, or largely, by their content of amino acids, particularly of those nine or ten
amino acids commonly classed as dietary essentials on the evidence of Rose’s well-known experiments on growing rats. Undoubtedly the content of a food in those amino acids that the body cannot synthesize from ordinary dietary components seta an upper limit to its usefulness in serving the biological functions of dietary protein; and to this extent the basis is sound. The amino acid analysis of the proteins in a food product Ly chemical methods is laborious and the various analyses for individual amino acids are not of equal precision; those for leucine and valine in particular leave much to be desired ( 4 ) . The microbiological methods suggested for determining the amino acid contents of food products are still in the experimental stage. However, it is not too much to hope that in the near future satisfactory methods, chemical or microbiological in nature, will be available for all the amino acids commonly believed to be dietary essentials. I t should be emphasized, however, that satisfactory analyses for all these essential amino acids must be a t hand before any one food product can be evaluated; otherwise, the possibility exists that the amino acid, or acids, for which’ no analysis is available may be, or include, the amino acid limiting the nutritive value of the contained proteins. When the ultimate goal of amino acid analysis of food proteins is attained, it will be possible from the data secured on many foods to predict which are the best in supplying the needs of the body for the essential amino acids, in so far as these needs are known, and in particular to predict which protein mixtures in individual foods will correct best each other’s deficiencies in e% sential amino acids. But these predictions cannot be expressed quantitatively nor can they be made with any great assurance, because the chemical picture of food proteins is inevitably altered and distorted by biological factors before a true picture of nutritional quality is secured. The disturbing biological factors that impair the usefulness in practical nutrition of a knowledge of the amino acid contents of foods are as follows: The digestibility of food proteins iu the alimentary tract IS largely independent of amino acid composition, and may be determined, as Mendcl and Fine showed many years ago (16-19) by the constituents of foods other than protein. The di eetive apparatus of the animal, though remarkably efficient unler the most favorable conditions in reducing dietary protein to its ultimate building stones, may be impeded and frustrated in its operation by those nonprotein constituents of food that successfully resist its attack (the cellulosic and hemicellulosic components in particular), while preventing complete acceea of the proteolytic enzymes to their pro er substrates. The indigestible residue of rood protein may not be represent. tive in its amino acid constitution of the food from which it wae derived. This may be inferred from the known differences in the ease with which different peptide linkages are split by the digestive enzymes. An instance in point was reported by Jon- and
November, 1944
ANALYTICAL EDITION
Waterman (IS)in their study of the digeation in vitro of arachin, the chief protein of the peanut. The amino acids are libcrated from food protein in digestion, not evenly, but a t different rates characteristic of the amino acid, or of the manner of its linka e in the protein molecule ( 1 2 , 29). They are also absorbed into &e blood a t diffcrent rates (6) The fate of the end roducts of protein digestion in metabolism 18 not determined s o l e t by their suitability in meeting the needs of the tissues for anabolic purposes. The tissues are also continually in need of energy in the performance of physiological work. The organic nutrients serve interchangeably in the metabolic mixture from which this energy is extracted. Amino acids resultin from protein di estion are thus inevitably drawn into the oxifative rcnctions of the body in proportion to their rclative concentrations in the influx to the tissues of the end products of digestion (21, 84, and probably also in proportion to their relative susccptibility to oxidation with reference to each other and to that of sugars and lipids. I n view of thc factors listed above, it would be remarkable indeed if the mixture of amino acids that is ultimately available to the tissues for synthesis of protcins and other nitrogenoub tissue constituents wrre closely similar to the mixture of amino arids present in the food aa consumed. Deficiencies in the lztter would of course persist throu h all these vicissitudes, but excwer of certain amino ncids maygbe wiped out, while pro ortions of others that were oiiginally adequate may develop into &ficiencies. I t seems fair to conclude that the amino acid analysis of food proteins may yield information of great significance concerning their nutritive values and their interrelationships in the diet, but that such a n analysis cannot take the place of a biological assay. It is doubtful that a n amino acid analysis of foods can measure, or explain, the changes in nutritive value that occur during storage and heat processing, especially a n improvement in nutritive value such as many legume prot,eins undergo on heat treatment. QROWTH-PROMOTING VALUE OF FOOD PROTEINS
Of the biological assays of food proteins in current use, the most popular undoubtedly is the determination with young albino rats of the ratio of gain in body weight to protein consumed during a 4- to &week period on diets containing variable concentrations of protein such that, ordinarily, the concentratiorl wed is inadequate to promote maximal growth, being otherwise complete. The method is a simplification of one proposed 2.5 years ago by Osborne, Mendel, and Ferry (33). The original method proposed that food proteins be compared by determining the maximum ratio of gain in weight to protein consumed among ratios secured by varying the concentration of protein in the experimental diet. This maximum ratio for casein was 2.25 a t a protein level of 12%, and for lactalbumin, 3.01 a t a protein level of 7.9%. The simplified method in common use has served a useful purpose in comparisons of protein “quality” in the biological sense among many food products studied. For many purposes, it is probably satisfactory, though its most appealing characteristic is its simplicity. Requiring no control of tho food intake of the experimental animals and no other measurements than the periodical taking of body weights, it seems to be an ideal technique when comparisons of large numbers of food products are to be made. The method was subjerted to critical scrutiny in a paper published by the author 20 years ago (23). I n the present connection it may be well to restate the implications of the method, but to discuss them niainly in the light of information revealed since the publication of this review. In crediting dietary protein only with the growth iiiduced in experimental animals and in permitting an unlimited consurnption of food, the method implies that there is no requirement of protein for the mere niainlenaiice of life, since only on such an assumption would it be expected that a constan) ratio of gain in weight to protein eaten would be obtained regardless of the intake of protein. The reality of this implication would be difficult to defend a t the prebcnt rime, especially i n view o f O*horne
697
and Mendel’s determinations (39) of maintenance requirementa for various wheat proteins and the more extensive work of Ymuta (36). If the implication is wrong, one would expect the ratio of gain in weight to protein consumed to increase on the same diet as the intake of protein (food) increases, since with the greater intake of protein, a greater proportion of it would be available for growth. Hoagland and Snider (9) found this to be true in that inale rats, consuming larger amounts of food than female rata and consequently growing faster, exhibited almost always the larger ratio of gain to protein eaten. Stewart and associates (36) found that a restriction of the food intake of rats lowered considerably the ratio of gain to protein eaten for the same protein source. Thus, for rolled oats the ratio averaged 1.52 under conditions of ad libitum feeding, but only 1.14 when the protein (food) consumption was restricted by about 40%, Of the same significance is the fact that the ratio was almost twice as variable with unrestricted feeding as with equalized feeding, the standard deviations being, respectively, 0.167 and 0.094. The method of measuring protein quality by an efficiency ratio of growth to protein eaten implies that the protein content of the gains in body weight of growing animals is constant regardless of the age or size of animal, the quality of the protein, or the rate of growth. To the extent that the gains differ in their content of [rotein, fat, and water they do not represent equal nutritive effects, and hence are not comparable. For example, it would be expected that a gram of dietary protein would induce a greater gain in body weight of a n animal if this gain contained 15% of protein than if it contained 20%, assuming that other growth essentials are present in adequate amounts. I n scientific inveatk gations with farm animals it is generally recognized that live weight increase is not a reliable measure of nutritive effect, a point discussed by Armsby (2, pp. 196-9) as early as 1917. Since then evidence of the variable composition of live weight gains has been presented by Mitchell and Carman (26) for rats, and by Fraps and Carlyle (7) for chickens. Hamilton (8) hae shown that the live weight gains of growing rats, when put on at the same rate, may vary in their content of protein and energy with varying levels of the same dietary protein, while Kik (24) proved that two proteins differing in nutritive value and fed in such amounts to growing rats receiving equal amounts of total food as to induce equal gains in body weight, may produce different proportions of protein in the body weight gains, the better protein producing the higher content of this nutrient in the gain@ secured. There is a distinct tendency for the more rapid gains in body weight to have the greater content of fat and the smaller content of protein, as Mitchell (2s)has demonstrated for cattle arid Ellis and Zeller (6) for pigs. The disturbing effect of a variable composition of body weight gains in the interpretation of a protein nutrition cxpcrinient i p well illustrated by an experiment reported from t,he author’s 1:iboratory (3) concerned with the comparative nutritive value of the protein of milk curd and of the cheeses produced from it bv various types of ripening. When fed in equal amounts in airedfeeding tests, the protcin of Limburger cheese was founct) to be equal to that of fresh milk curd in growth-promoting value, although it waa definitely less digestible and possessed a lower biologicd value. The explanation lay in the composition of thr body weight gains: the gains put on by the Limburger cheese ration were definitely lower in protein content and higher in their fat content, than the gains produced on the milk-rurd ration. In the same sort of tests, Swiss cheese protein provrd superior to milk curd in growth-promoting value, though no better in bioiogjcal value and definitely inferior in digestibility. An aimlogous diftercnce in the composition of body weight gains between t,hr cliccsc ration and the milk-curd ration must have occurred. Thus, the two clear implications of this method for the bioirrgical assay of protein quality in nutrition are untenable. Some evidence of the seriousness of these basic errors i n producing variation in the ratio of body weight gain to prot,cin intake has been given above. The sit,uation may be further pictured by t.he hypothetical illustrationa givrn in Table I, showiiig to w h i t , ex-
I N D U S T R I A L A N D ’ E N G I N E E R I N G CHEMISTRY
698
Eliect of Increarin Intake of Same Diet on Ratio of Body Weight &in to Protein Intake Case 1 Case 2 Case 3 Case 4 Case 5 50 50 Weight of rat, grams 50 50 50 5 6 7 4 8 Daily food, g r a m 000 700 800 Daily protein, mg. 400 500 200 Protein for maintenance, mg. 200 200 200 200 500 400 000 Protein fo? growth, mg. 200 300 275 105 220 330 Net protein for growth, mg. 110 20 21 19 22 23 Protein content of gain, % 1.05 1.37 1.74 Dailv cain. crams 0.48 0.75 Table
1.
Gain p e r ’ i r a m of protein eaten, grams
1.20
1.50
1.75
1.90
2.17
tent the ratio may be changed by changing food intake, although the utilization of the dietary protein remains unaltered. In each of the five illustrations, the weight of rat is the same, but the daily consumption of food increases from 4 to 8 grams. The diet in all cases is the same and contains 10% of whole wheat protein. The protein requirement for maintenance for this protein was found by Osborne and Mendel (%) to approximate 3 mg. per gram of rat per day. This requirement has been raised to 4 mg. because at a 10% level in the diet, utilization in metabolism is less (21, 24). The wheat protein in all cases is assumed to be !5% digestible and to have a biological value of 65% ( 2 7 ) , the net protein” thus amounting to 55% (0.85 X 0.65 = 0.55) of the total protein. The protein content of the body weight gains is assumed to decrease from 23 to 19% as the food intake and the rate of growth increase, in conformance with evidence cited above. These assumptions seem reasonable and to a large extent are based upon experimental evidence. They lead to the expectation that the ratio of body weight gain to protein eaten may be changed, for a whole wheat diet containing 10% of protein, from 1.20 to 2.17 by merely increasing the intake of food. This demonstration accounts for the large experimental error to which the ratio is subjected under conditions of ad libitum feeding. I t also indicates clearly that this method of assaying protein quality will definitely tend to exaggerate differences in this respect among different protein foods, because the rations containing the better protein foods.wil1 generally be consumed in the larger amounts and will thus be given an undue advantage over the poorer protein foods. Equating the food intakes of rats on comparable diets will overcome this effect, but the uncertainty concerning the protein content of body weight gains is inherent in the method. In brief, this method of assaying food protein biologically may be expected to exaggerate quality differences among proteins when these are considerable, and to obscure them when they are inconsiderable, on account of the large experimental error to which the method is subject as it is ordinarily employed. BIOLOGICAL ASSAY OF FOOD PROTEINS BY MEANS OF NITROGEN METABOLISM STUDIES
Vol. 16, No. 11
mines the storage of protein in growth rather than assumes that this storage is proportional to body weight gains. Furthermore, its accuracy is such (86)that it can detect differences in digestibility and biological value of proteins of a magnitude of two or three percentage units using a battery of only 10 animals. An illustration of its value in detecting changes in the nutritive value of a food protein (soybean proteins) during storage is afforded by the data summarized in Table 11. The soybeans used in this test were of the Illini variety and the
1942 crop. They were stored at a temperature ef 78” to 80’ F.
in air-ti ht containers, either whole or as a ground defatted meal that h a 3 been heated in the autoclave for 1.5 hours at 17 pounds steam pressure. After 8.5 months, and again after 12 months’ storage, containers of both whole beans and autoclaved meal were opened and tested for protein quality by the nitrogen balance method. The whole beans before testing were ground, extracted with ether, and heated for 1.5 hours in the autoclave at 17 ounds’ pressure-Le., the same treatment to which the meal haabeen subjected prior to storage. The results indicate that after 8.5 months’ storage, the proteins in the beans stored whole with no pretreatment had suffered a marked decrease in digestibility, averaging 7 percentage units, and an even more marked decrease in biological value, averaging 11 percentage units, as compared with the beans ground and heated rior to storage. Both differences were statistically significant. i t the end of 12 months’ storage, the difference in digestibility between the two products was still distinct, but a clear difference in biological value was not demonstrated, possibly because of two erratic biological values secured for the ground meal. The results confirm and extend those reported by Jones and Gersdorff (If), who reported a drop in the digestibility in vitro of the proteins of soybeans on storage. The results of a similar test on corn proteins are summarized in Table 111. In this test the corn was stored either whole or ground, but with no preheating. The method of storing evidently had no effect, either on the digestibility or the biological value of the corn proteins, but storage for 8 months depressed equally the biological value of the proteins in corn stored whole or ground, while leaving the digestibility unaffected. From this outcome, one might conclude that, in the soybean test, it was the preheating, rather than the grinding, that exerted the stabilizing effect on the proteins during storage. The results on corn do not harmonize entirely with those reported by Jones, Divine, and Gersdorff (10). NET PROTEIN VALUE OF FOOD PRODUCTS
While the nitrogen balance method for the biological assay of protein quality distinguishes betn-een losses of nitrogen in the digestive and in the metabolic processes of the animal body, there are distinct advantages in securing in a single figure an appraisal of the value of a food as a source of dietary protein. Such a figure should involve not only the biological value of the protein and its digestibility, hut also its original content of “conventional”
This method was first introduced by Thomas (37) in 1909, and was later adaDted to the arowina - rat by Mitchell (20). Xlany modifications in the method have since been made (25, 28). I t is based upon the fact that protein is the only considerable source of dietary nitrogen and that the disposiTable 11. Changes in Digestibility and Biological Value of Proteins tion of protein within the body of an animal can be followed of Soybeans during Storage with considerable accuracy by determining the intake and 1.411 determinations made on beans after autoclavine) _. Soybeans Stored 12 Months Soybeans Stored 8.5 Months fecal and urinary excretion of nitrogen. When this is done Whole Ground, Wbole, Ground, under carefully selected and controlled conditions, it is Unheatdd Autoclaved Unheated Autoclaved .~ possible to compute from the data secured ( a ) the digestiTrue RioTrue BioEioTrue BjoTrue Rat digest- logical digest- logical Rat digest- logical digest- logical bility of the protein, ( b ) its utilization in metabolism for No. ibilitv value ibilitv value No. ibility value ibility value all purposes, expressed as a percentage called by Thomas “0 % %% % % % % 337 78 60 .. 65 85 80 292 77 the “biological value”, and ( c ) the “net protein content” 85 71 340 79 67 61 83 71 295 78 of the food assayed, equal to the total content times the 82 82 343 80 67 61 86 73 298 76 85 I31 84 70 346 77 70 67 .. 301 77 _ _ _. digestion coefficient (expressed as a decimal) times the 83 52 349 81 66 71 85 304 84 61 85 73 79 338 78 72 293 82 64 88 biological value (also expressed as a decimal). 84 67 68 341 81 74 65 85 296 79 Unlike the biological assay method previously con86 68 344 79 71 62 59 84 299 75 347 81 76 84 76 60 60 87 302 76 sidered, this method credits the dietary protein with all 82 350 79 07 63 69 86 79 305 79 its characteristic functions in the body, maintenance as 79.4 66.1 84.0 08.2 Av. 78.3 02.6 85.3 73.7 well as growth in the growing rat, and it directly deter~~
~
November, 1944
protein-i.e., N X 6.25. Such a value was proposed by the author and Carman (27) many years ago, and has been called the “net protein value” of the food which may be computed on the dry basis or on the fresh basis, as in the tabulation below:
Food Eggs Lean ham Wheat Soybeans Soybeans, stored for 8 5 months
699
ANALYTICAL EDITION
Total Protein
Dikestibility of Protein
Digestible Protein
Biological Value
Net Protein
%
%
%
13.4 19.8 12.5 32.8
100 100 91 84
% 13.4 19.8 11.4 27.6
93 74 67 72
% 12.5 14.6 7.6 19 9
32.8
78
25.6
63
16.1
The net protein values as t,hus computed are mainly of comparative significance, because of the variation to which biological values are subject as the level of protein in the diet changes. CRITIQUE OF THE NITROGEN BALANCE METHOD OF ASSAY OF FOOD PROTEIN
Measurements of protein digestibility and of the biological value of the absorbed protein must, in the present state of biochemical technique, be expressed in terms of nitrogen. This is not the most satisfactory situation, because all food nitrogen is not in the form of protein, or amino acids, or amino acid derivatives, nor is the nitrogen of the tissues present there only in such compounds as these. However, the ambiguity introduced into measurements of protein digestibility and biological value by employing nitrogen as a conventional equivalent of protein is not generally large. Certainly no substitute procedure can claim to be as satisfactory. Perhaps a more serious objection to this emphasis upon protein nitrogen is that it neglects the function of dietary protein of providing the body with comparatively large amounts of readily 2.Vailable phosphorus, which is not an integral part of the amino acid structure. The nitrogen balance method of assessing protein quality was pioposed by Thomas a t a time when Folin’s theory of protein metabolism, with its sharp distinction between exogenous and endogenous metabolism, was commonly accepted. In assuming a constant (biologically speaking) erosion of nitrogenous material from the animal body, commensurate with the maintenance requirement of protein, the method tacitly assumes the reality of Folin’s endogenous catabolism. However, the work of Schoenheimer and his associates on amino acid metabolism using the nitrogen isotope may seem to cast doubt upon any assumption of this nature. These revolutionary investigations swept away all sharp distinctions between the endogenous and the exogenous metabolism that form the basis of Folin’s theory of protein metabolism. They afford a convincing picture of the dynamic equilibrium existing between the tissue proteins and the amino acids of dietary origin in the circulating fluids of the body. However, they do not change the significance of the end results of protein metabolism as revealed by such a nitrogen balance study as would be undertaken in the course of securing digestion coefficients and biological values according to the Thomas method or some modification of it. The nitrogen in the urine on a nitrogenfree diet still represents a minimum or basal level of nitrogen loss from the tissues, bearing a relationship to the basal metabolism of energy (35). Considerable circumstantial evidence can be cited to the effect that this basal rate of nitrogen output in the urine continues a t a constant level, biologically speaking, during regimes of adequate protein nutrition. One of the most convincing bits of evidence of this description may be found in a publication by Ackerson and Blish ( I ) on the utilization of the protein of corn by hens. Nonmolting hens excreted an average of 143 mg. of nitrogen per kg. of body weight per day on a nitrogen-free diet, while molting hens on the same diet excreted 217 mg. However, assuming that these greatly different values for the constant catabolism of nitrogen persisted during periods of feeding a corn diet, the same average biological
value of 68 wm obtained for a group of 19 nonmolting birds and for a group of 8 molting birds. The urinary nitrogen during periods when test proteins are being fed consists definitely of two fractions, one the constant contribution from the tissues, and the other related to the level of protein feeding and the efficiency with which the absorbed amino acids satisfy the prevailing requirements of amino acids by the tissues. This fraction is partly of tissue origin and partly of dietary origin, because of the dynamic relation existing between tissue proteins and dietary amino acids. But this dynamic relationship is not anarchistic in character. Through the welter of reactions in which the tissue proteins are involved, the cells and organs of the adult body retain their original form and size and the chemical structures of the protein molecules emerge unchanged, as Schoenheimer and Rittenberg admit (34). Discussing the many interconversion reactions divulged by the isotope method, Moss and Schoenheimer (SO) believe that they cannot be regarded “merely as steps in metabolic degradation. They seem rather to represent automatic and noninterruptable biochemical processes, of synthesis as well as degradation, which are balanced by an unknown regulatory mechanism so that the total amount of the body material and its composition do not change.” Table Ill. Changes in Digestibility and Biological Value of Protein (Nitrogen) of Corn (U. S. Hybrid 13) during Storage After Storage for 8 Months Rat No.
Before Storage True Biodigestlogical ibility value
% 259 261 263 265 267 260 262 264 266 268 Av.
90 91 93 94 93 90 89 95 92 96 92.3
% 72 62 69 68 69 62 61 55 62 69 64.9
Rat
No.
Stored Whole True Biodigestlogical ibility value
%
%
91 92 93 91 89 92 33 93 92 91 91.7
53 54 57 56 52 60 67 58 61 59 57.7
Stored Ground True Biodigestlogical ibility value
% 92 92 94 91 90 91 90 91 91 91 91.3
% 60 60 60
55 62 60
59 54 56 60
58.6
The “exogenous nitrogen” of the urine, to use the obsolete term of Folin, is thus both of tissue and of diet origin, but the rate of its excretion can nevertheless be assumed to depend solely upon the level of protein intake and upon the extent to which the dietary protein is used in the replacement of tissue losses of nitrogen in the adult animal. I n the growing animal the “exogenous nitrogen” of the urine will be reduced to the extent that dietary amino acids are used in the elaboration of new tissue in growth. To all intents and purposes, therefore, this fraction of the urine nitrogen measures the wastage of dietary nitrogen in metabolism under the conditions of diet control imposed in a well-considered determination of the biological value of protein. The essential accuracy of the assumptions upon which this determination is based has been tested in the author’s laboratory by comparison with a method of measuring protein quality that does not involve these assumptions. The results obtained in tests upon two protein sources were found to agree satisfaetorily (26). The reproducibility of results secured a t different times upon the same protein source is also satisfactory as the method is used in this laboratory, although very occasionally unaccountable erratic values, such as three of those listed in Table 11, are obtained. Any method of biological assay of food products may thus go “haywire” on occasion. NITROGEN REPLACEMENT VALUFS
Murlin and his associates (31) have introduced a new method of interpreting nitrogen balance data relating to the utilization
m
INDUSTRIAL A N D ENGINEERING CHEMISTRY T8ble
IV. Replacement Values of Protein of Exploded Wheat. on Proteins of Unprocessed Wheatb
Period I.
Daily Nitrogen Dail Nitrogen Intake Jalance Ex- UnprocEx- UnprocPair Rat ploded essed Aver- ploded essed DifferNo. No. wheat wheat age wheat wheat ences MQ. Ma. IUQ. MQ. MQ. MQ.
p%.proteili 1
in &eta
2 3 4 K
11,49 protein 1 in Jiete 2 3 4 K
A?. b
1 2 3 4 6
e
7 8 9
10
1
2 3 4 6 6 7 8 9 10
150
iii iSi ...
134
iii ...
...
88
...
79
... 62 ...
... 169
... 143 ,,,
142
,..
142
,..
142 87 ... 7a , .
61 .. 89
90
...
96
.. .
...
85
.. . 155 ... ...
-3.5
....
-8.9
RGplacement Value8
.. . . . . . . 17.5 20.8 .... .... 9.3 18.2 . ...
... 78 .._
-17.0
777-81 (1939).
...
.... .. . . -26.5 61 -34.6 .... ... . . . . -7.9 89-31.6 .... ... , . . . -18.2 96 -29.1 .... -34:s
88
17.8
77
9.1
86 ... 74
.... ....
23.6
,...
10.9
315-21 (1934).
Ellis, N. R., and Zeller, J. H., U. S. Dept. Agr., Tech. Fraps, G. 9.. and Carlyle, E. C., J . A g r . Research, 59
86
10.K
....
Beadles, J. R., Quieenberry, J. H.;Nakamura, F. I., and Mitchell, H. H., J . Agr. Research, 47, 947 (1933). Block, R. J., and Bolling, D., “Amino Acid Composition of Proteins and Natural Foods. Analytical Methods and Results”, Springfield, Ill., C. C. Thomas, 1944. Chase, B. W., and Lewis, H. B., J. Biol. C h m . . 106. Bull. 413 (1934).
% ...
Hamilton, T. So,J. Nzctritiop, 17, 565-82 (1939). Hoagland, R., and Snider, G. G., J . AQT.Research, 32
.... 87 15.4 138 .... 6.4 Lio:o io;’ ... -6.8 .... .... 138 .... ..., 12.1 17.9 87’ -14.4 .... .. . ii8 . . . . 6.9 20.3 86 . . . . . . . -16.8 . . . . . . 87 -26.3 .... 138
Vol. 16, No. 11
1025-40 (1926).
Jones, D. B., Divine, J. P., and Gersdorff, C. E. F . Cereal Chem., 19,’819-30 (1942).
Jones, D. B., and Gersdorff, C. E. F., J. Am. C h m . Soc., 60. 723 119381.
Jones, D. B., anh Gersdorff, C. E. F., J . Biol.Chem.,
.. .
101
657-67 (1933).
...
Jones, D. B., and Waterman, H. C., Ibid., 52,357 (1922) Kik. M. C., Arkansas Anr. ExDt. Sta., Bull. 352 (1938). Mendel, L. B., and Fine, M-, S., J. Biol. Chem.. 10
...
303-25 (191 1-12) Ibid., 10, 339-43 (1911-12). Ibid., 10, 345-52 (1911-12). Ibid., 10, 433-58 (1911-12). Ibid., 11, 1-3, 5-26 (1912). Mitchell, H . H., J. B b l . C h m . , 58, 873 (1924). Ibid., 58, 905-22 (1924). Mitchell. H . H.. Natl. Res. Council. Bull. 67 (1928). . . Mitchell, H. H,’,Physiil. Reus., 4, 424-78 (1924). Mitchell, H. H., and Beadles, J. R., J . Bid. Chem., 71, 429 (1927). Mitchell, H. H., Burrougha, E. W., and Beadles, J. R., J . Nutrition, 11, 257-274 (1936). Mitchell, H. H., and Carman, G. G., Am. J. Physiol., 76, 89s410 (1926). Mitchell, H. H., and Carman, G. G., J. Biol. Chem.. 60, 613 (1924). Ibii.: 6 8 , .183-2 15 ( 1926).
89 86.8
“Puffed Wheat”, Quaker Oats Co. Test carned out with aid of funds from Qeneral Mills,h a .
of protein in nutrition that posseases the advantage of removing the need of a standardizing period for the determination of the body’s contribution to the fecal and urinary nitrogen. It was devised especially for use in experiments on human subjects, for whom low-protein diets are extremely unpalatable. The method simply compares the nitrogen balances of adult experimental subjects on the same intake of nitrogen from the food under test and in another period, from a reference protein of high nutritive value, such as the protein of milk or of egg. The method may be applied to such a problem as that of the injury to food protein brought about by heat processing. In Table l V , the method has been used to assess the extent of heat injury induced in the whole wheat kernel by subjecting it to the “gun explosion’’ method of preparing a breakfast food. Five pairs of adult rats have been carried through two nitrogen metabolism periods, one rat iqeach pair receiving its protein from proceased wheat and the other rat in equal amounts from unproceased wheat. In the second metabolism period the sources of protein for the paired rats were reversed. The replacement values given in the last column of the table indicate the extent of heat injury and were computed by the formula
Mitchell, H. H., and Hamilton, T. S., “Biochemistry of the Amino Acids”, A.C.S. MonouraDh No. 48, New York, Reinhold Publishing Corp., 19%. Moss, A. R., and Schoenheimer, R., J. Biol. Chem., 135, 416 (1940).
Murlin, J. R., Nasset, E. B., and Marsh, M. E., J. Nutritia. 16, 249-69 (1938).
Osborne, T. B., and Mendel, L. B., J . Biol. Chem., 37, 557-601 (1919).
Osborne, T. B., Mendel, L. B., and Ferry, E. L., Ibid., 37, 2239 (1919).
Schoenheimer, R., and Rittenberp, D., Physiol. Reus., 20,
218
(1940). Smuts, D. B., J. Nutrition.,9, 403-33 (1935).
Stewart, R. A., Hensley, G. W., and Peters, F. N., Jr., Ibid.. 26, 519-26 (1943).
Thomas,, Karl, Arch. Am.&. Phyeiol., Physiol. A b t . , 219-302 (1909).
in which Bi is the nitrogen balance of the rat subsisting on the unprocessed wheat diet, BI is the nitrogen balance of its pair mate subsisting on the processed wheat diet, and I is.the average of the nitrogen intakes of the two rats in the pair, which should be practically the same. The average replacement value of 86.8 shows that the processing of wheat by the gun explosion method has impaired the protein value in adult rodent nutrition by 13.2y0. For growing animals, the impairment may be greater than this. It is interesting t o note that, with adult human subjects, Murlin, Nasset, and Marsh (32) report egg replacement values of 70 and 57 for two whole wheat cereals (Ralston’s Wheat Cereal and Puffed Wheat) subjected, respectively, to mild procwing and to the extreme heat of the gun explosion process. The latter value is 81.4% of the former, equivalent to a heat injury to the contained protein of 18,6y0. LITERATURE CITED (1)
Ackerson, C. W., and Bliah, M. J., Poultru Sci., 5,
226-32
(1926). (2) Armshy, H. P., “Nutrition of Farm Animals”, p. 743, New York, .Maomillan Co., 1917.
TEIS article represents a complete revinion for publication purposes of a paper presented before the Division of Agricultural and Food Chemistry a .the 107th Meeting of t h e AM~ERICAN CHEMICAL SOCIETY, Cleveland, Obi,.
WPB Restrictions on Laboratory Equipment WPB reduced the number of types of laboratory equipmont that may be sold or delivered only upon authorization by amendment to Order L-144, issued October 22. Types of laboratory equipment which are in short supply sre still subject to control: analytical balances (sensitivity 0.05 mg. or more sensitive) : centrifuges valued at more than 880 each; hydrogen-ion meters, electrometric type; metalloscopes and metallographs; microscopes, stereoscopic wide field; Abbe refractometers; spectrogrspha (quarts), spectrophotometers (quarts), and spectrornetera (infrared); and vacuum pumps (one micron or higher vacuum).
,
